Using Keras for building a DNN¶
In [1]:
import tensorflow as tf
import matplotlib.pyplot as plt
import os
from pandas import read_csv
from sklearn.model_selection import train_test_split
%matplotlib inline
1. Load and prepare the data¶
In [2]:
myfile = 'diamond_prices.csv'
diamonds = read_csv(myfile)
diamonds.head(10)
Out[2]:
| carat | cut | color | clarity | depth | table | price | x | y | z | |
|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0.23 | Ideal | E | SI2 | 61.5 | 55.0 | 326 | 3.95 | 3.98 | 2.43 |
| 1 | 0.21 | Premium | E | SI1 | 59.8 | 61.0 | 326 | 3.89 | 3.84 | 2.31 |
| 2 | 0.23 | Good | E | VS1 | 56.9 | 65.0 | 327 | 4.05 | 4.07 | 2.31 |
| 3 | 0.29 | Premium | I | VS2 | 62.4 | 58.0 | 334 | 4.20 | 4.23 | 2.63 |
| 4 | 0.31 | Good | J | SI2 | 63.3 | 58.0 | 335 | 4.34 | 4.35 | 2.75 |
| 5 | 0.24 | Very Good | J | VVS2 | 62.8 | 57.0 | 336 | 3.94 | 3.96 | 2.48 |
| 6 | 0.24 | Very Good | I | VVS1 | 62.3 | 57.0 | 336 | 3.95 | 3.98 | 2.47 |
| 7 | 0.26 | Very Good | H | SI1 | 61.9 | 55.0 | 337 | 4.07 | 4.11 | 2.53 |
| 8 | 0.22 | Fair | E | VS2 | 65.1 | 61.0 | 337 | 3.87 | 3.78 | 2.49 |
| 9 | 0.23 | Very Good | H | VS1 | 59.4 | 61.0 | 338 | 4.00 | 4.05 | 2.39 |
Splice the data to get predictive feature, feature vectors¶
In [3]:
from sklearn.preprocessing import LabelEncoder
In [4]:
encoder = LabelEncoder()
diamonds["cut"] = encoder.fit_transform(diamonds["cut"])
diamonds["color"] = encoder.fit_transform(diamonds["color"])
diamonds["clarity"] = encoder.fit_transform(diamonds["clarity"])
In [5]:
TARGET = 'price'
X_data = diamonds.iloc[:,1:].values
y_data = diamonds[TARGET].values
In [6]:
X_data
Out[6]:
array([[2. , 1. , 3. , ..., 3.95, 3.98, 2.43],
[3. , 1. , 2. , ..., 3.89, 3.84, 2.31],
[1. , 1. , 4. , ..., 4.05, 4.07, 2.31],
...,
[4. , 0. , 2. , ..., 5.66, 5.68, 3.56],
[3. , 4. , 3. , ..., 6.15, 6.12, 3.74],
[2. , 0. , 3. , ..., 5.83, 5.87, 3.64]])
In [7]:
y_data
Out[7]:
array([ 326, 326, 327, ..., 2757, 2757, 2757], dtype=int64)
2. Splitting Data into training and testing sets¶
In [8]:
X_train, X_test, y_train, y_test = train_test_split(X_data, y_data, test_size=0.1, random_state=60)
Number of Inputs¶
In [9]:
n_inputs = X_train.shape[1]
3. Build the DL model using Dense layers¶
In [10]:
model = tf.keras.models.Sequential()
# first hidden layer, you only need to set the input_dim for the first layer
model.add(tf.keras.layers.Dense(units=128, activation='relu', input_dim=n_inputs))
# second hidden layer
model.add(tf.keras.layers.Dense(units=64, activation='relu'))
# third hidden layer
model.add(tf.keras.layers.Dense(units=32, activation='relu'))
# output layer # for activation: If you don't specify anything, no activation is applied
model.add(tf.keras.layers.Dense(units=1))
4. Compile the model¶
In [11]:
model.compile(loss='mean_squared_error',
optimizer='adam',
metrics=['mean_squared_error'])
5. Train the model¶
In [12]:
N_EPOCHS = 400
BATCH_SIZE = 128
model.fit(X_train, y_train, epochs=N_EPOCHS, batch_size=BATCH_SIZE)
Epoch 1/400 48546/48546 [==============================] - 2s 41us/step - loss: 854870.2177 - mean_squared_error: 854870.2177 Epoch 2/400 48546/48546 [==============================] - 1s 22us/step - loss: 5.1288 - mean_squared_error: 5.1288 Epoch 3/400 48546/48546 [==============================] - 1s 22us/step - loss: 0.9146 - mean_squared_error: 0.9146 Epoch 4/400 48546/48546 [==============================] - 1s 22us/step - loss: 1.3673 - mean_squared_error: 1.3673 Epoch 5/400 48546/48546 [==============================] - 1s 23us/step - loss: 6.3234 - mean_squared_error: 6.3234 Epoch 6/400 48546/48546 [==============================] - 1s 22us/step - loss: 0.1119 - mean_squared_error: 0.1119 Epoch 7/400 48546/48546 [==============================] - 1s 23us/step - loss: 1.3100 - mean_squared_error: 1.3100 Epoch 8/400 48546/48546 [==============================] - 1s 22us/step - loss: 5.2690 - mean_squared_error: 5.2690 Epoch 9/400 48546/48546 [==============================] - 1s 22us/step - loss: 0.1008 - mean_squared_error: 0.1008 Epoch 10/400 48546/48546 [==============================] - 1s 22us/step - loss: 0.1075 - mean_squared_error: 0.1075 Epoch 11/400 48546/48546 [==============================] - 1s 22us/step - loss: 2.2121 - mean_squared_error: 2.2121 Epoch 12/400 48546/48546 [==============================] - 1s 22us/step - loss: 3.0033 - mean_squared_error: 3.0033 Epoch 13/400 48546/48546 [==============================] - 1s 22us/step - loss: 1.2537 - mean_squared_error: 1.2537 Epoch 14/400 48546/48546 [==============================] - 1s 22us/step - loss: 10.8334 - mean_squared_error: 10.8334 Epoch 15/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.8328 - mean_squared_error: 0.8328 Epoch 16/400 48546/48546 [==============================] - 1s 22us/step - loss: 0.2642 - mean_squared_error: 0.2642 Epoch 17/400 48546/48546 [==============================] - 1s 22us/step - loss: 92.3564 - mean_squared_error: 92.3564 Epoch 18/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.7428 - mean_squared_error: 0.7428 Epoch 19/400 48546/48546 [==============================] - 1s 27us/step - loss: 1065.4473 - mean_squared_error: 1065.4473 Epoch 20/400 48546/48546 [==============================] - 1s 29us/step - loss: 89.8901 - mean_squared_error: 89.8901 Epoch 21/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.1516 - mean_squared_error: 0.1516 Epoch 22/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.2276 - mean_squared_error: 0.2276 Epoch 23/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.2208 - mean_squared_error: 0.2208 Epoch 24/400 48546/48546 [==============================] - 1s 23us/step - loss: 14.0615 - mean_squared_error: 14.0615 Epoch 25/400 48546/48546 [==============================] - 1s 24us/step - loss: 157.2437 - mean_squared_error: 157.2437 Epoch 26/400 48546/48546 [==============================] - 1s 23us/step - loss: 70.3666 - mean_squared_error: 70.3666 Epoch 27/400 48546/48546 [==============================] - 1s 23us/step - loss: 5.3282 - mean_squared_error: 5.3282 Epoch 28/400 48546/48546 [==============================] - 1s 24us/step - loss: 329.5448 - mean_squared_error: 329.5448 Epoch 29/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.1733 - mean_squared_error: 0.1733 Epoch 30/400 48546/48546 [==============================] - 1s 25us/step - loss: 1.0335 - mean_squared_error: 1.0335 1s Epoch 31/400 48546/48546 [==============================] - 1s 24us/step - loss: 315.4548 - mean_squared_error: 315.4548 Epoch 32/400 48546/48546 [==============================] - 1s 24us/step - loss: 1.0494 - mean_squared_error: 1.0494 Epoch 33/400 48546/48546 [==============================] - 1s 24us/step - loss: 22.0806 - mean_squared_error: 22.0806 Epoch 34/400 48546/48546 [==============================] - 1s 24us/step - loss: 159.5661 - mean_squared_error: 159.5661 Epoch 35/400 48546/48546 [==============================] - 1s 24us/step - loss: 1.6696 - mean_squared_error: 1.6696 Epoch 36/400 48546/48546 [==============================] - 1s 24us/step - loss: 196.8387 - mean_squared_error: 196.8387 Epoch 37/400 48546/48546 [==============================] - 1s 24us/step - loss: 296.6857 - mean_squared_error: 296.6857 Epoch 38/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.1469 - mean_squared_error: 0.1469 Epoch 39/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.3247 - mean_squared_error: 0.3247 Epoch 40/400 48546/48546 [==============================] - 1s 25us/step - loss: 497.8490 - mean_squared_error: 497.8490 Epoch 41/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.1347 - mean_squared_error: 0.1347 Epoch 42/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.1131 - mean_squared_error: 0.1131 Epoch 43/400 48546/48546 [==============================] - 1s 28us/step - loss: 0.1902 - mean_squared_error: 0.1902 Epoch 44/400 48546/48546 [==============================] - 1s 26us/step - loss: 298.0079 - mean_squared_error: 298.0079 Epoch 45/400 48546/48546 [==============================] - 1s 27us/step - loss: 0.0935 - mean_squared_error: 0.0935 Epoch 46/400 48546/48546 [==============================] - 1s 26us/step - loss: 0.2172 - mean_squared_error: 0.2172 Epoch 47/400 48546/48546 [==============================] - 1s 26us/step - loss: 1520.7537 - mean_squared_error: 1520.7537 Epoch 48/400 48546/48546 [==============================] - 1s 26us/step - loss: 0.0495 - mean_squared_error: 0.0495 Epoch 49/400 48546/48546 [==============================] - 1s 27us/step - loss: 0.0478 - mean_squared_error: 0.0478 Epoch 50/400 48546/48546 [==============================] - 1s 27us/step - loss: 0.0485 - mean_squared_error: 0.0485 Epoch 51/400 48546/48546 [==============================] - 1s 27us/step - loss: 0.0558 - mean_squared_error: 0.0558 Epoch 52/400 48546/48546 [==============================] - 1s 28us/step - loss: 0.1856 - mean_squared_error: 0.1856 Epoch 53/400 48546/48546 [==============================] - 1s 24us/step - loss: 5.3831 - mean_squared_error: 5.3831 Epoch 54/400 48546/48546 [==============================] - 1s 25us/step - loss: 105.9065 - mean_squared_error: 105.9065 Epoch 55/400 48546/48546 [==============================] - 1s 23us/step - loss: 121.5662 - mean_squared_error: 121.5662 Epoch 56/400 48546/48546 [==============================] - 1s 23us/step - loss: 4.4790 - mean_squared_error: 4.4790 Epoch 57/400 48546/48546 [==============================] - 1s 23us/step - loss: 92.3632 - mean_squared_error: 92.3632 Epoch 58/400 48546/48546 [==============================] - 1s 23us/step - loss: 20.2658 - mean_squared_error: 20.2658 Epoch 59/400 48546/48546 [==============================] - 1s 24us/step - loss: 108.8890 - mean_squared_error: 108.8890 Epoch 60/400 48546/48546 [==============================] - 1s 23us/step - loss: 432.6360 - mean_squared_error: 432.6360 Epoch 61/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.0521 - mean_squared_error: 0.0521 Epoch 62/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0429 - mean_squared_error: 0.0429 Epoch 63/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.6492 - mean_squared_error: 0.6492 Epoch 64/400 48546/48546 [==============================] - 1s 24us/step - loss: 151.1430 - mean_squared_error: 151.1430 Epoch 65/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.0448 - mean_squared_error: 0.0448 Epoch 66/400 48546/48546 [==============================] - 1s 26us/step - loss: 217.8201 - mean_squared_error: 217.8201 Epoch 67/400 48546/48546 [==============================] - 1s 27us/step - loss: 56.2015 - mean_squared_error: 56.2015 Epoch 68/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0730 - mean_squared_error: 0.0730 Epoch 69/400 48546/48546 [==============================] - 1s 23us/step - loss: 34.9165 - mean_squared_error: 34.9165 Epoch 70/400 48546/48546 [==============================] - 1s 24us/step - loss: 16.0748 - mean_squared_error: 16.0748 Epoch 71/400 48546/48546 [==============================] - 1s 22us/step - loss: 557.7442 - mean_squared_error: 557.7442 Epoch 72/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.0197 - mean_squared_error: 0.0197 Epoch 73/400 48546/48546 [==============================] - 1s 22us/step - loss: 0.0298 - mean_squared_error: 0.0298 Epoch 74/400 48546/48546 [==============================] - 1s 22us/step - loss: 1.3228 - mean_squared_error: 1.3228 Epoch 75/400 48546/48546 [==============================] - 1s 23us/step - loss: 293.8951 - mean_squared_error: 293.8951 Epoch 76/400 48546/48546 [==============================] - 1s 22us/step - loss: 0.0340 - mean_squared_error: 0.0340 Epoch 77/400 48546/48546 [==============================] - 1s 22us/step - loss: 0.0372 - mean_squared_error: 0.0372 Epoch 78/400 48546/48546 [==============================] - 1s 23us/step - loss: 381.9286 - mean_squared_error: 381.9286 Epoch 79/400 48546/48546 [==============================] - 1s 22us/step - loss: 149.4718 - mean_squared_error: 149.4718 Epoch 80/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.0170 - mean_squared_error: 0.0170 Epoch 81/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.0137 - mean_squared_error: 0.0137 Epoch 82/400 48546/48546 [==============================] - 1s 22us/step - loss: 0.1873 - mean_squared_error: 0.1873 Epoch 83/400 48546/48546 [==============================] - 1s 23us/step - loss: 179.4953 - mean_squared_error: 179.4953 Epoch 84/400 48546/48546 [==============================] - 1s 22us/step - loss: 0.0186 - mean_squared_error: 0.0186 Epoch 85/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.8800 - mean_squared_error: 0.8800 Epoch 86/400 48546/48546 [==============================] - 1s 23us/step - loss: 125.5541 - mean_squared_error: 125.5541 Epoch 87/400 48546/48546 [==============================] - 1s 22us/step - loss: 0.2643 - mean_squared_error: 0.2643 Epoch 88/400 48546/48546 [==============================] - 1s 22us/step - loss: 179.8670 - mean_squared_error: 179.8670 Epoch 89/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.1407 - mean_squared_error: 0.1407 Epoch 90/400 48546/48546 [==============================] - 1s 22us/step - loss: 191.7731 - mean_squared_error: 191.7731 Epoch 91/400 48546/48546 [==============================] - 1s 22us/step - loss: 0.0859 - mean_squared_error: 0.0859 Epoch 92/400 48546/48546 [==============================] - 1s 22us/step - loss: 0.3858 - mean_squared_error: 0.3858 Epoch 93/400 48546/48546 [==============================] - 1s 25us/step - loss: 387.5378 - mean_squared_error: 387.5378 Epoch 94/400 48546/48546 [==============================] - 1s 29us/step - loss: 0.0116 - mean_squared_error: 0.0116 Epoch 95/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0150 - mean_squared_error: 0.0150 Epoch 96/400 48546/48546 [==============================] - 2s 32us/step - loss: 0.6605 - mean_squared_error: 0.6605 Epoch 97/400 48546/48546 [==============================] - 1s 24us/step - loss: 212.0133 - mean_squared_error: 212.0133 Epoch 98/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0152 - mean_squared_error: 0.0152 Epoch 99/400 48546/48546 [==============================] - 1s 25us/step - loss: 9.5142 - mean_squared_error: 9.5142 Epoch 100/400 48546/48546 [==============================] - 1s 24us/step - loss: 98.4579 - mean_squared_error: 98.4579 Epoch 101/400 48546/48546 [==============================] - 1s 25us/step - loss: 37.1011 - mean_squared_error: 37.1011 Epoch 102/400 48546/48546 [==============================] - 1s 25us/step - loss: 32.8912 - mean_squared_error: 32.8912 Epoch 103/400 48546/48546 [==============================] - 1s 24us/step - loss: 252.9903 - mean_squared_error: 252.9903 Epoch 104/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0116 - mean_squared_error: 0.0116 Epoch 105/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0395 - mean_squared_error: 0.0395 Epoch 106/400 48546/48546 [==============================] - 1s 26us/step - loss: 120.6768 - mean_squared_error: 120.6768 Epoch 107/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0362 - mean_squared_error: 0.0362 Epoch 108/400 48546/48546 [==============================] - 1s 25us/step - loss: 48.2808 - mean_squared_error: 48.2808 Epoch 109/400 48546/48546 [==============================] - 1s 27us/step - loss: 151.2373 - mean_squared_error: 151.2373 Epoch 110/400 48546/48546 [==============================] - 1s 25us/step - loss: 45.0314 - mean_squared_error: 45.0314 Epoch 111/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0319 - mean_squared_error: 0.0319 Epoch 112/400 48546/48546 [==============================] - 1s 24us/step - loss: 257.0849 - mean_squared_error: 257.0849 Epoch 113/400 48546/48546 [==============================] - 1s 25us/step - loss: 58.3225 - mean_squared_error: 58.3225 Epoch 114/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0052 - mean_squared_error: 0.0052 Epoch 115/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0079 - mean_squared_error: 0.0079 Epoch 116/400 48546/48546 [==============================] - 1s 24us/step - loss: 49.6675 - mean_squared_error: 49.6675 Epoch 117/400 48546/48546 [==============================] - 1s 24us/step - loss: 285.0253 - mean_squared_error: 285.0253 Epoch 118/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.7030 - mean_squared_error: 0.7030 Epoch 119/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0066 - mean_squared_error: 0.0066 Epoch 120/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0210 - mean_squared_error: 0.0210 Epoch 121/400 48546/48546 [==============================] - 1s 24us/step - loss: 135.0525 - mean_squared_error: 135.0525 Epoch 122/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.1528 - mean_squared_error: 0.1528 Epoch 123/400 48546/48546 [==============================] - 1s 24us/step - loss: 6.6564 - mean_squared_error: 6.6564 Epoch 124/400 48546/48546 [==============================] - 1s 24us/step - loss: 67.5603 - mean_squared_error: 67.5603 Epoch 125/400 48546/48546 [==============================] - 1s 24us/step - loss: 83.3606 - mean_squared_error: 83.3606 Epoch 126/400 48546/48546 [==============================] - 1s 25us/step - loss: 53.6833 - mean_squared_error: 53.6833A: 0s - loss: 88.4724 - mean_sq Epoch 127/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.9275 - mean_squared_error: 0.9275 1s - Epoch 128/400 48546/48546 [==============================] - 1s 25us/step - loss: 160.1887 - mean_squared_error: 160.1887 Epoch 129/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0841 - mean_squared_error: 0.0841 Epoch 130/400 48546/48546 [==============================] - 1s 24us/step - loss: 159.9586 - mean_squared_error: 159.9586 Epoch 131/400 48546/48546 [==============================] - 1s 25us/step - loss: 6.5074 - mean_squared_error: 6.5074 Epoch 132/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.1162 - mean_squared_error: 0.1162 Epoch 133/400 48546/48546 [==============================] - 1s 24us/step - loss: 95.1971 - mean_squared_error: 95.1971 Epoch 134/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.4931 - mean_squared_error: 0.4931 Epoch 135/400 48546/48546 [==============================] - 1s 25us/step - loss: 86.3506 - mean_squared_error: 86.3506 Epoch 136/400 48546/48546 [==============================] - 1s 24us/step - loss: 1.8889 - mean_squared_error: 1.8889 Epoch 137/400 48546/48546 [==============================] - 1s 24us/step - loss: 80.6957 - mean_squared_error: 80.6957 Epoch 138/400 48546/48546 [==============================] - 1s 23us/step - loss: 24.8143 - mean_squared_error: 24.8143 Epoch 139/400 48546/48546 [==============================] - 1s 24us/step - loss: 261.0190 - mean_squared_error: 261.0190 Epoch 140/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.0021 - mean_squared_error: 0.0021 Epoch 141/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.0047 - mean_squared_error: 0.0047 Epoch 142/400 48546/48546 [==============================] - 1s 23us/step - loss: 8.4646 - mean_squared_error: 8.4646 Epoch 143/400 48546/48546 [==============================] - 1s 25us/step - loss: 91.4612 - mean_squared_error: 91.4612 Epoch 144/400 48546/48546 [==============================] - 1s 24us/step - loss: 45.6921 - mean_squared_error: 45.6921 Epoch 145/400 48546/48546 [==============================] - 1s 28us/step - loss: 40.6186 - mean_squared_error: 40.6186 Epoch 146/400 48546/48546 [==============================] - 1s 24us/step - loss: 103.3741 - mean_squared_error: 103.3741 Epoch 147/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.1385 - mean_squared_error: 0.1385 Epoch 148/400 48546/48546 [==============================] - 1s 24us/step - loss: 179.9796 - mean_squared_error: 179.9796 Epoch 149/400 48546/48546 [==============================] - 1s 24us/step - loss: 45.7151 - mean_squared_error: 45.7151 Epoch 150/400 48546/48546 [==============================] - 1s 26us/step - loss: 0.0019 - mean_squared_error: 0.0019 Epoch 151/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0175 - mean_squared_error: 0.0175 Epoch 152/400 48546/48546 [==============================] - 1s 25us/step - loss: 90.9734 - mean_squared_error: 90.9734 Epoch 153/400 48546/48546 [==============================] - 1s 26us/step - loss: 0.0772 - mean_squared_error: 0.0772 Epoch 154/400 48546/48546 [==============================] - 1s 26us/step - loss: 92.8861 - mean_squared_error: 92.8861 Epoch 155/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0405 - mean_squared_error: 0.0405 Epoch 156/400 48546/48546 [==============================] - 1s 26us/step - loss: 248.9959 - mean_squared_error: 248.9959 Epoch 157/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0021 - mean_squared_error: 0.0021 Epoch 158/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0096 - mean_squared_error: 0.0096 Epoch 159/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0711 - mean_squared_error: 0.0711 Epoch 160/400 48546/48546 [==============================] - 1s 24us/step - loss: 131.1459 - mean_squared_error: 131.1459 Epoch 161/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0110 - mean_squared_error: 0.0110 Epoch 162/400 48546/48546 [==============================] - 1s 24us/step - loss: 26.2457 - mean_squared_error: 26.2457 Epoch 163/400 48546/48546 [==============================] - 1s 24us/step - loss: 177.9387 - mean_squared_error: 177.9387 Epoch 164/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0025 - mean_squared_error: 0.0025 Epoch 165/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.7416 - mean_squared_error: 0.7416 Epoch 166/400 48546/48546 [==============================] - 2s 43us/step - loss: 181.0100 - mean_squared_error: 181.0100 Epoch 167/400 48546/48546 [==============================] - 2s 34us/step - loss: 0.0065 - mean_squared_error: 0.0065 Epoch 168/400 48546/48546 [==============================] - 2s 32us/step - loss: 0.8988 - mean_squared_error: 0.8988 Epoch 169/400 48546/48546 [==============================] - 2s 37us/step - loss: 72.8411 - mean_squared_error: 72.8411 Epoch 170/400 48546/48546 [==============================] - 2s 32us/step - loss: 116.1649 - mean_squared_error: 116.1649 Epoch 171/400 48546/48546 [==============================] - 1s 31us/step - loss: 4.2027e-04 - mean_squared_error: 4.2027e-04 0s - loss: 2.6644e Epoch 172/400 48546/48546 [==============================] - 1s 27us/step - loss: 56.9909 - mean_squared_error: 56.9909 Epoch 173/400 48546/48546 [==============================] - 1s 25us/step - loss: 71.5946 - mean_squared_error: 71.5946 Epoch 174/400 48546/48546 [==============================] - 1s 26us/step - loss: 0.0034 - mean_squared_error: 0.0034 Epoch 175/400 48546/48546 [==============================] - 1s 25us/step - loss: 102.1147 - mean_squared_error: 102.1147 Epoch 176/400 48546/48546 [==============================] - 1s 28us/step - loss: 5.4485 - mean_squared_error: 5.4485 Epoch 177/400 48546/48546 [==============================] - 1s 28us/step - loss: 444.8648 - mean_squared_error: 444.8648 Epoch 178/400 48546/48546 [==============================] - 1s 29us/step - loss: 5.8720 - mean_squared_error: 5.8720 Epoch 179/400 48546/48546 [==============================] - 1s 29us/step - loss: 0.0012 - mean_squared_error: 0.0012 Epoch 180/400 48546/48546 [==============================] - 1s 30us/step - loss: 0.0011 - mean_squared_error: 0.0011 Epoch 181/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0048 - mean_squared_error: 0.0048 Epoch 182/400 48546/48546 [==============================] - 1s 25us/step - loss: 7.2004 - mean_squared_error: 7.2004 Epoch 183/400 48546/48546 [==============================] - 1s 24us/step - loss: 132.7935 - mean_squared_error: 132.7935 Epoch 184/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0319 - mean_squared_error: 0.0319 Epoch 185/400 48546/48546 [==============================] - 1s 24us/step - loss: 217.6532 - mean_squared_error: 217.6532 Epoch 186/400 48546/48546 [==============================] - 1s 25us/step - loss: 22.6350 - mean_squared_error: 22.6350 Epoch 187/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0017 - mean_squared_error: 0.0017 Epoch 188/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0044 - mean_squared_error: 0.0044 Epoch 189/400 48546/48546 [==============================] - 1s 25us/step - loss: 118.5524 - mean_squared_error: 118.5524 Epoch 190/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0043 - mean_squared_error: 0.0043 Epoch 191/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.2386 - mean_squared_error: 0.2386 Epoch 192/400 48546/48546 [==============================] - 1s 24us/step - loss: 148.5231 - mean_squared_error: 148.5231 Epoch 193/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0069 - mean_squared_error: 0.0069 Epoch 194/400 48546/48546 [==============================] - 1s 25us/step - loss: 94.9057 - mean_squared_error: 94.9057 Epoch 195/400 48546/48546 [==============================] - 1s 30us/step - loss: 0.0206 - mean_squared_error: 0.0206 Epoch 196/400 48546/48546 [==============================] - ETA: 0s - loss: 145.4948 - mean_squared_error: 145.49 - 1s 24us/step - loss: 170.6654 - mean_squared_error: 170.6654 Epoch 197/400 48546/48546 [==============================] - 1s 24us/step - loss: 47.6729 - mean_squared_error: 47.6729 Epoch 198/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0020 - mean_squared_error: 0.0020 Epoch 199/400 48546/48546 [==============================] - 1s 24us/step - loss: 1.7966 - mean_squared_error: 1.7966 Epoch 200/400 48546/48546 [==============================] - 1s 24us/step - loss: 128.1034 - mean_squared_error: 128.1034 Epoch 201/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0033 - mean_squared_error: 0.0033 Epoch 202/400 48546/48546 [==============================] - 1s 26us/step - loss: 0.9863 - mean_squared_error: 0.9863 Epoch 203/400 48546/48546 [==============================] - 1s 25us/step - loss: 52.7798 - mean_squared_error: 52.7798 Epoch 204/400 48546/48546 [==============================] - 1s 25us/step - loss: 281.8574 - mean_squared_error: 281.8574 Epoch 205/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0020 - mean_squared_error: 0.0020 Epoch 206/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.0021 - mean_squared_error: 0.0021 Epoch 207/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.0322 - mean_squared_error: 0.0322 Epoch 208/400 48546/48546 [==============================] - 1s 23us/step - loss: 756.1894 - mean_squared_error: 756.1894 Epoch 209/400 48546/48546 [==============================] - 1s 23us/step - loss: 136.0292 - mean_squared_error: 136.0292 Epoch 210/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.0027 - mean_squared_error: 0.0027 Epoch 211/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.0022 - mean_squared_error: 0.0022 Epoch 212/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.0026 - mean_squared_error: 0.0026 Epoch 213/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.0017 - mean_squared_error: 0.0017 Epoch 214/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0183 - mean_squared_error: 0.0183 Epoch 215/400 48546/48546 [==============================] - 1s 23us/step - loss: 97.9471 - mean_squared_error: 97.9471 Epoch 216/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.0064 - mean_squared_error: 0.0064 Epoch 217/400 48546/48546 [==============================] - 1s 23us/step - loss: 385.1228 - mean_squared_error: 385.1228 Epoch 218/400 48546/48546 [==============================] - 1s 23us/step - loss: 12.7142 - mean_squared_error: 12.7142 Epoch 219/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.0013 - mean_squared_error: 0.0013 Epoch 220/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0015 - mean_squared_error: 0.0015 Epoch 221/400 48546/48546 [==============================] - 1s 23us/step - loss: 0.0036 - mean_squared_error: 0.0036 Epoch 222/400 48546/48546 [==============================] - 1s 25us/step - loss: 93.6371 - mean_squared_error: 93.6371 Epoch 223/400 48546/48546 [==============================] - 1s 26us/step - loss: 0.0028 - mean_squared_error: 0.0028 Epoch 224/400 48546/48546 [==============================] - 1s 29us/step - loss: 198.6759 - mean_squared_error: 198.6759 Epoch 225/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0024 - mean_squared_error: 0.0024 Epoch 226/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0014 - mean_squared_error: 0.0014 Epoch 227/400 48546/48546 [==============================] - 1s 25us/step - loss: 182.8279 - mean_squared_error: 182.8279 Epoch 228/400 48546/48546 [==============================] - 1s 26us/step - loss: 0.0127 - mean_squared_error: 0.0127 Epoch 229/400 48546/48546 [==============================] - 1s 26us/step - loss: 0.0038 - mean_squared_error: 0.0038 Epoch 230/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0165 - mean_squared_error: 0.0165 Epoch 231/400 48546/48546 [==============================] - 1s 26us/step - loss: 114.9985 - mean_squared_error: 114.9985 Epoch 232/400 48546/48546 [==============================] - 1s 26us/step - loss: 0.0450 - mean_squared_error: 0.0450 Epoch 233/400 48546/48546 [==============================] - 1s 26us/step - loss: 339.1311 - mean_squared_error: 339.1311 Epoch 234/400 48546/48546 [==============================] - 1s 28us/step - loss: 0.0225 - mean_squared_error: 0.0225 Epoch 235/400 48546/48546 [==============================] - ETA: 0s - loss: 8.4247e-04 - mean_squared_error: 8.4247e- - 1s 25us/step - loss: 8.3707e-04 - mean_squared_error: 8.3707e-04 Epoch 236/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0012 - mean_squared_error: 0.0012 Epoch 237/400 48546/48546 [==============================] - 1s 26us/step - loss: 1.8799 - mean_squared_error: 1.8799 Epoch 238/400 48546/48546 [==============================] - 1s 28us/step - loss: 72.7178 - mean_squared_error: 72.7178 Epoch 239/400 48546/48546 [==============================] - 1s 26us/step - loss: 0.0377 - mean_squared_error: 0.0377 Epoch 240/400 48546/48546 [==============================] - 1s 30us/step - loss: 158.3671 - mean_squared_error: 158.3671 Epoch 241/400 48546/48546 [==============================] - 2s 38us/step - loss: 9.5869e-04 - mean_squared_error: 9.5869e-04 Epoch 242/400 48546/48546 [==============================] - 1s 29us/step - loss: 0.0141 - mean_squared_error: 0.0141 Epoch 243/400 48546/48546 [==============================] - ETA: 0s - loss: 241.4393 - mean_squared_error: 241.43 - 1s 25us/step - loss: 236.1776 - mean_squared_error: 236.1776 Epoch 244/400 48546/48546 [==============================] - 1s 25us/step - loss: 5.6685e-04 - mean_squared_error: 5.6685e-04 Epoch 245/400 48546/48546 [==============================] - 1s 25us/step - loss: 9.1781e-04 - mean_squared_error: 9.1781e-04 Epoch 246/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0065 - mean_squared_error: 0.0065 Epoch 247/400 48546/48546 [==============================] - 1s 30us/step - loss: 32.1308 - mean_squared_error: 32.1308 Epoch 248/400 48546/48546 [==============================] - 1s 25us/step - loss: 96.3884 - mean_squared_error: 96.3884 Epoch 249/400 48546/48546 [==============================] - 1s 26us/step - loss: 0.0225 - mean_squared_error: 0.0225 Epoch 250/400 48546/48546 [==============================] - 1s 24us/step - loss: 10.5910 - mean_squared_error: 10.5910 Epoch 251/400 48546/48546 [==============================] - 1s 26us/step - loss: 141.0666 - mean_squared_error: 141.0666 Epoch 252/400 48546/48546 [==============================] - 1s 26us/step - loss: 0.0021 - mean_squared_error: 0.0021 Epoch 253/400 48546/48546 [==============================] - 1s 24us/step - loss: 9.9969 - mean_squared_error: 9.9969 Epoch 254/400 48546/48546 [==============================] - 1s 24us/step - loss: 45.1106 - mean_squared_error: 45.1106 Epoch 255/400 48546/48546 [==============================] - 1s 24us/step - loss: 300.1683 - mean_squared_error: 300.1683 Epoch 256/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0277 - mean_squared_error: 0.0277 Epoch 257/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0013 - mean_squared_error: 0.0013 Epoch 258/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0030 - mean_squared_error: 0.0030 Epoch 259/400 48546/48546 [==============================] - 1s 25us/step - loss: 107.3068 - mean_squared_error: 107.3068 Epoch 260/400 48546/48546 [==============================] - 1s 25us/step - loss: 1.1877 - mean_squared_error: 1.1877 Epoch 261/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.1240 - mean_squared_error: 0.1240 Epoch 262/400 48546/48546 [==============================] - 1s 24us/step - loss: 42.4701 - mean_squared_error: 42.4701 Epoch 263/400 48546/48546 [==============================] - 1s 24us/step - loss: 100.0143 - mean_squared_error: 100.0143 Epoch 264/400 48546/48546 [==============================] - 1s 26us/step - loss: 0.0924 - mean_squared_error: 0.0924 Epoch 265/400 48546/48546 [==============================] - 1s 24us/step - loss: 149.2910 - mean_squared_error: 149.2910 Epoch 266/400 48546/48546 [==============================] - 1s 25us/step - loss: 2.9806 - mean_squared_error: 2.9806 Epoch 267/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0040 - mean_squared_error: 0.0040 Epoch 268/400 48546/48546 [==============================] - 1s 24us/step - loss: 4.6653 - mean_squared_error: 4.6653 Epoch 269/400 48546/48546 [==============================] - 1s 31us/step - loss: 56.6712 - mean_squared_error: 56.6712 Epoch 270/400 48546/48546 [==============================] - 2s 32us/step - loss: 45.1483 - mean_squared_error: 45.1483 Epoch 271/400 48546/48546 [==============================] - 2s 32us/step - loss: 18.6607 - mean_squared_error: 18.6607 Epoch 272/400 48546/48546 [==============================] - 1s 30us/step - loss: 103.6307 - mean_squared_error: 103.6307 Epoch 273/400 48546/48546 [==============================] - 1s 27us/step - loss: 17.6039 - mean_squared_error: 17.6039 Epoch 274/400 48546/48546 [==============================] - 1s 26us/step - loss: 0.1856 - mean_squared_error: 0.1856 Epoch 275/400 48546/48546 [==============================] - 1s 28us/step - loss: 88.7213 - mean_squared_error: 88.7213 Epoch 276/400 48546/48546 [==============================] - 1s 27us/step - loss: 0.0267 - mean_squared_error: 0.0267 Epoch 277/400 48546/48546 [==============================] - 2s 34us/step - loss: 190.8842 - mean_squared_error: 190.8842 Epoch 278/400 48546/48546 [==============================] - 1s 25us/step - loss: 6.2893 - mean_squared_error: 6.2893 Epoch 279/400 48546/48546 [==============================] - 1s 30us/step - loss: 0.0028 - mean_squared_error: 0.0028 Epoch 280/400 48546/48546 [==============================] - 1s 28us/step - loss: 0.0042 - mean_squared_error: 0.0042 Epoch 281/400 48546/48546 [==============================] - 1s 25us/step - loss: 130.2085 - mean_squared_error: 130.2085 Epoch 282/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0021 - mean_squared_error: 0.0021 Epoch 283/400 48546/48546 [==============================] - 1s 29us/step - loss: 0.1676 - mean_squared_error: 0.1676 Epoch 284/400 48546/48546 [==============================] - 1s 26us/step - loss: 640.5174 - mean_squared_error: 640.5174 Epoch 285/400 48546/48546 [==============================] - 1s 27us/step - loss: 0.0024 - mean_squared_error: 0.0024 Epoch 286/400 48546/48546 [==============================] - 2s 37us/step - loss: 0.0017 - mean_squared_error: 0.0017 Epoch 287/400 48546/48546 [==============================] - 2s 32us/step - loss: 0.0016 - mean_squared_error: 0.0016 Epoch 288/400 48546/48546 [==============================] - 2s 32us/step - loss: 0.0024 - mean_squared_error: 0.0024 1s - l Epoch 289/400 48546/48546 [==============================] - 1s 27us/step - loss: 0.0027 - mean_squared_error: 0.0027 Epoch 290/400 48546/48546 [==============================] - 1s 26us/step - loss: 30.6015 - mean_squared_error: 30.6015 Epoch 291/400 48546/48546 [==============================] - 1s 25us/step - loss: 7.1580 - mean_squared_error: 7.1580 Epoch 292/400 48546/48546 [==============================] - 1s 25us/step - loss: 185.5546 - mean_squared_error: 185.5546 Epoch 293/400 48546/48546 [==============================] - 1s 26us/step - loss: 0.0046 - mean_squared_error: 0.0046 Epoch 294/400 48546/48546 [==============================] - 1s 26us/step - loss: 0.0356 - mean_squared_error: 0.0356 Epoch 295/400 48546/48546 [==============================] - 1s 26us/step - loss: 0.3462 - mean_squared_error: 0.3462 Epoch 296/400 48546/48546 [==============================] - 1s 25us/step - loss: 218.9785 - mean_squared_error: 218.9785 Epoch 297/400 48546/48546 [==============================] - 1s 26us/step - loss: 0.0046 - mean_squared_error: 0.0046 Epoch 298/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0078 - mean_squared_error: 0.0078 Epoch 299/400 48546/48546 [==============================] - 1s 24us/step - loss: 22.1699 - mean_squared_error: 22.1699 Epoch 300/400 48546/48546 [==============================] - 1s 25us/step - loss: 36.7277 - mean_squared_error: 36.7277 Epoch 301/400 48546/48546 [==============================] - 1s 25us/step - loss: 26.7427 - mean_squared_error: 26.7427 Epoch 302/400 48546/48546 [==============================] - 1s 25us/step - loss: 39.8256 - mean_squared_error: 39.8256 Epoch 303/400 48546/48546 [==============================] - 1s 25us/step - loss: 2.3291 - mean_squared_error: 2.3291 Epoch 304/400 48546/48546 [==============================] - 1s 25us/step - loss: 62.8201 - mean_squared_error: 62.8201 Epoch 305/400 48546/48546 [==============================] - 1s 29us/step - loss: 85.7280 - mean_squared_error: 85.7280 Epoch 306/400 48546/48546 [==============================] - 1s 27us/step - loss: 39.0092 - mean_squared_error: 39.0092 Epoch 307/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0087 - mean_squared_error: 0.0087 Epoch 308/400 48546/48546 [==============================] - 1s 25us/step - loss: 66.3307 - mean_squared_error: 66.3307 Epoch 309/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.1275 - mean_squared_error: 0.1275 Epoch 310/400 48546/48546 [==============================] - 1s 26us/step - loss: 119.4176 - mean_squared_error: 119.4176 Epoch 311/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.6269 - mean_squared_error: 0.6269 Epoch 312/400 48546/48546 [==============================] - 1s 29us/step - loss: 0.0072 - mean_squared_error: 0.0072 Epoch 313/400 48546/48546 [==============================] - 1s 30us/step - loss: 183.3674 - mean_squared_error: 183.3674 Epoch 314/400 48546/48546 [==============================] - 2s 42us/step - loss: 0.8527 - mean_squared_error: 0.8527 Epoch 315/400 48546/48546 [==============================] - 2s 35us/step - loss: 0.0041 - mean_squared_error: 0.0041 Epoch 316/400 48546/48546 [==============================] - 2s 31us/step - loss: 0.0035 - mean_squared_error: 0.0035 Epoch 317/400 48546/48546 [==============================] - 2s 35us/step - loss: 36.2909 - mean_squared_error: 36.2909 Epoch 318/400 48546/48546 [==============================] - 2s 35us/step - loss: 27.6123 - mean_squared_error: 27.6123 Epoch 319/400 48546/48546 [==============================] - 1s 28us/step - loss: 181.0274 - mean_squared_error: 181.0274 Epoch 320/400 48546/48546 [==============================] - 1s 29us/step - loss: 0.0016 - mean_squared_error: 0.0016 Epoch 321/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0043 - mean_squared_error: 0.0043 Epoch 322/400 48546/48546 [==============================] - 1s 28us/step - loss: 226.7263 - mean_squared_error: 226.7263 Epoch 323/400 48546/48546 [==============================] - 1s 29us/step - loss: 0.0157 - mean_squared_error: 0.0157 Epoch 324/400 48546/48546 [==============================] - 2s 37us/step - loss: 0.0012 - mean_squared_error: 0.0012 Epoch 325/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0029 - mean_squared_error: 0.0029 Epoch 326/400 48546/48546 [==============================] - 1s 29us/step - loss: 0.0643 - mean_squared_error: 0.0643 Epoch 327/400 48546/48546 [==============================] - 1s 29us/step - loss: 194.6298 - mean_squared_error: 194.6298 Epoch 328/400 48546/48546 [==============================] - 1s 30us/step - loss: 6.9672e-04 - mean_squared_error: 6.9672e-04 Epoch 329/400 48546/48546 [==============================] - 1s 31us/step - loss: 0.0019 - mean_squared_error: 0.0019 Epoch 330/400 48546/48546 [==============================] - 1s 30us/step - loss: 38.1798 - mean_squared_error: 38.1798 Epoch 331/400 48546/48546 [==============================] - 1s 31us/step - loss: 26.7784 - mean_squared_error: 26.7784 Epoch 332/400 48546/48546 [==============================] - 1s 29us/step - loss: 32.9106 - mean_squared_error: 32.9106 Epoch 333/400 48546/48546 [==============================] - 1s 28us/step - loss: 2.8765 - mean_squared_error: 2.8765 Epoch 334/400 48546/48546 [==============================] - 1s 30us/step - loss: 103.7213 - mean_squared_error: 103.7213 Epoch 335/400 48546/48546 [==============================] - 1s 30us/step - loss: 0.0058 - mean_squared_error: 0.0058 Epoch 336/400 48546/48546 [==============================] - 1s 29us/step - loss: 29.6365 - mean_squared_error: 29.6365 Epoch 337/400 48546/48546 [==============================] - 1s 27us/step - loss: 22.5337 - mean_squared_error: 22.5337 Epoch 338/400 48546/48546 [==============================] - 1s 28us/step - loss: 220.7363 - mean_squared_error: 220.7363 Epoch 339/400 48546/48546 [==============================] - 1s 29us/step - loss: 0.0026 - mean_squared_error: 0.0026 Epoch 340/400 48546/48546 [==============================] - 1s 29us/step - loss: 0.0020 - mean_squared_error: 0.0020 Epoch 341/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0023 - mean_squared_error: 0.0023 Epoch 342/400 48546/48546 [==============================] - 1s 24us/step - loss: 62.7673 - mean_squared_error: 62.7673 Epoch 343/400 48546/48546 [==============================] - 1s 24us/step - loss: 33.5770 - mean_squared_error: 33.5770 Epoch 344/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0183 - mean_squared_error: 0.0183 Epoch 345/400 48546/48546 [==============================] - 1s 25us/step - loss: 130.8831 - mean_squared_error: 130.8831 Epoch 346/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0018 - mean_squared_error: 0.0018 Epoch 347/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0023 - mean_squared_error: 0.0023 Epoch 348/400 48546/48546 [==============================] - 1s 24us/step - loss: 124.4147 - mean_squared_error: 124.4147 Epoch 349/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0019 - mean_squared_error: 0.0019 Epoch 350/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0268 - mean_squared_error: 0.0268 Epoch 351/400 48546/48546 [==============================] - 1s 24us/step - loss: 70.7843 - mean_squared_error: 70.7843 Epoch 352/400 48546/48546 [==============================] - 1s 25us/step - loss: 2.5766 - mean_squared_error: 2.5766 Epoch 353/400 48546/48546 [==============================] - 1s 25us/step - loss: 13.9382 - mean_squared_error: 13.9382 Epoch 354/400 48546/48546 [==============================] - 1s 25us/step - loss: 136.7479 - mean_squared_error: 136.7479 Epoch 355/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0042 - mean_squared_error: 0.0042 Epoch 356/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0851 - mean_squared_error: 0.0851 Epoch 357/400 48546/48546 [==============================] - 1s 26us/step - loss: 473.0768 - mean_squared_error: 473.0768 Epoch 358/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0014 - mean_squared_error: 0.0014 Epoch 359/400 48546/48546 [==============================] - 1s 25us/step - loss: 9.6885e-04 - mean_squared_error: 9.6885e-04 Epoch 360/400 48546/48546 [==============================] - ETA: 0s - loss: 7.2687e-04 - mean_squared_error: 7.2687e- - 1s 25us/step - loss: 7.2394e-04 - mean_squared_error: 7.2394e-04 Epoch 361/400 48546/48546 [==============================] - 1s 24us/step - loss: 6.7745e-04 - mean_squared_error: 6.7745e-04 Epoch 362/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0031 - mean_squared_error: 0.0031 Epoch 363/400 48546/48546 [==============================] - 1s 24us/step - loss: 248.0456 - mean_squared_error: 248.0456 Epoch 364/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.0252 - mean_squared_error: 0.0252 Epoch 365/400 48546/48546 [==============================] - 1s 26us/step - loss: 0.0012 - mean_squared_error: 0.0012 Epoch 366/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0011 - mean_squared_error: 0.0011 Epoch 367/400 48546/48546 [==============================] - 1s 24us/step - loss: 30.0827 - mean_squared_error: 30.0827 Epoch 368/400 48546/48546 [==============================] - 1s 25us/step - loss: 53.5655 - mean_squared_error: 53.5655 Epoch 369/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0052 - mean_squared_error: 0.0052 Epoch 370/400 48546/48546 [==============================] - 1s 24us/step - loss: 19.0474 - mean_squared_error: 19.0474 Epoch 371/400 48546/48546 [==============================] - 1s 24us/step - loss: 38.1083 - mean_squared_error: 38.1083 Epoch 372/400 48546/48546 [==============================] - 1s 24us/step - loss: 66.4654 - mean_squared_error: 66.4654 Epoch 373/400 48546/48546 [==============================] - 1s 25us/step - loss: 14.2404 - mean_squared_error: 14.2404 Epoch 374/400 48546/48546 [==============================] - 1s 26us/step - loss: 0.8439 - mean_squared_error: 0.8439 Epoch 375/400 48546/48546 [==============================] - 1s 25us/step - loss: 46.0595 - mean_squared_error: 46.0595 Epoch 376/400 48546/48546 [==============================] - 1s 27us/step - loss: 72.5870 - mean_squared_error: 72.5870 Epoch 377/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0092 - mean_squared_error: 0.0092 Epoch 378/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.0028 - mean_squared_error: 0.0028 Epoch 379/400 48546/48546 [==============================] - 1s 28us/step - loss: 116.1074 - mean_squared_error: 116.1074 Epoch 380/400 48546/48546 [==============================] - 1s 28us/step - loss: 0.0052 - mean_squared_error: 0.0052 Epoch 381/400 48546/48546 [==============================] - 1s 27us/step - loss: 4.4453 - mean_squared_error: 4.4453 Epoch 382/400 48546/48546 [==============================] - 1s 28us/step - loss: 48.7047 - mean_squared_error: 48.7047 Epoch 383/400 48546/48546 [==============================] - 1s 28us/step - loss: 17.4554 - mean_squared_error: 17.4554 Epoch 384/400 48546/48546 [==============================] - 1s 28us/step - loss: 235.8987 - mean_squared_error: 235.8987 Epoch 385/400 48546/48546 [==============================] - 1s 28us/step - loss: 0.0124 - mean_squared_error: 0.0124 Epoch 386/400 48546/48546 [==============================] - 1s 29us/step - loss: 9.3536e-04 - mean_squared_error: 9.3536e-04 Epoch 387/400 48546/48546 [==============================] - 1s 29us/step - loss: 0.0041 - mean_squared_error: 0.0041 Epoch 388/400 48546/48546 [==============================] - 1s 29us/step - loss: 3.1611 - mean_squared_error: 3.1611 0s - loss: 1.7938 - mean Epoch 389/400 48546/48546 [==============================] - 1s 27us/step - loss: 70.7901 - mean_squared_error: 70.7901 Epoch 390/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.3529 - mean_squared_error: 0.3529 Epoch 391/400 48546/48546 [==============================] - 1s 25us/step - loss: 66.5627 - mean_squared_error: 66.5627 Epoch 392/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.1179 - mean_squared_error: 0.1179 Epoch 393/400 48546/48546 [==============================] - 1s 25us/step - loss: 50.6662 - mean_squared_error: 50.6662 Epoch 394/400 48546/48546 [==============================] - 1s 24us/step - loss: 0.5648 - mean_squared_error: 0.5648 Epoch 395/400 48546/48546 [==============================] - 1s 24us/step - loss: 35.3852 - mean_squared_error: 35.3852 Epoch 396/400 48546/48546 [==============================] - 1s 25us/step - loss: 4.7387 - mean_squared_error: 4.7387 Epoch 397/400 48546/48546 [==============================] - 1s 24us/step - loss: 71.1123 - mean_squared_error: 71.1123 Epoch 398/400 48546/48546 [==============================] - 1s 25us/step - loss: 0.1701 - mean_squared_error: 0.1701 Epoch 399/400 48546/48546 [==============================] - 1s 26us/step - loss: 50.9312 - mean_squared_error: 50.9312 Epoch 400/400 48546/48546 [==============================] - 1s 25us/step - loss: 21.7911 - mean_squared_error: 21.7911
Out[12]:
<tensorflow.python.keras.callbacks.History at 0x249adfab518>
6. Visualize/analyze the results of the model¶
In [13]:
## Getting the predictions from the model
predictions = model.predict(X_test).flatten()
In [14]:
fig, ax = plt.subplots(figsize=(8,5))
ax.scatter(x=predictions, y=y_test, s=0.5)
ax.set_xlabel('Predicted prices')
ax.set_ylabel('Observed prices')
ax.set_title("Predictions vs. Observed Values in the validation set");
